Domain Authority Auditor

This skill evaluates domain authority across 40 standardized criteria organized in 4 dimensions. It produces a comprehensive audit report with per-item scoring, dimension and weighted scores by domain type, veto item checks, and a prioritized action plan.

Sister skill: content-quality-auditor evaluates content at the page level (80 items). This skill evaluates the domain behind the content (40 items). Together they provide a complete 120-item assessment.

Namespace note: CITE uses C01-C10 for Citation items; CORE-EEAT uses C01-C10 for Contextual Clarity items. In combined 120-item assessments, prefix with the framework name (e.g., CITE-C01 vs CORE-C01) to avoid confusion.

System role: Citation Trust Gate. It decides whether a domain is credible enough to support ranking, citation, and brand authority work.

When This Must Trigger

Use this when domain credibility or citation trustworthiness is in question — even if the user doesn't use audit terminology:

• User asks "how trustworthy is my site" or "is my domain credible"
• When backlink-analyzer finds toxic link ratio above 15%, its handoff summary recommends this gate check
• Evaluating domain authority before a GEO campaign
• Benchmarking your domain against competitors
• Assessing whether a domain is trustworthy as a citation source
• Running periodic domain health checks or after link building campaigns
• Identifying manipulation red flags (PBNs, link farms, penalty history)
• Cross-referencing with content-quality-auditor for full 120-item assessment

What This Skill Does

1. Full 40-Item Audit: Scores every CITE check item as Pass/Partial/Fail
2. Dimension Scoring: Calculates scores for all 4 dimensions (0-100 each)
3. Weighted Totals: Applies domain-type-specific weights for CITE Score
4. Critical Issue Detection: Flags critical manipulation signals that cap the score
5. Priority Ranking: Identifies Top 5 improvements sorted by impact
6. Action Plan: Generates specific, actionable improvement steps
7. Cross-Reference: Optionally pairs with CORE-EEAT for combined diagnosis

Quick Start

Start with one of these prompts. Finish with a citation-trust verdict and a handoff summary using the repository format in Skill Contract.

Audit Your Domain

Audit domain authority for [domain]
Run a CITE domain audit on [domain] as a [domain type]

Audit with Domain Type

CITE audit for example.com as an e-commerce site
Score this SaaS domain against the 40-item benchmark: [domain]

Comparative Audit

Compare domain authority: [your domain] vs [competitor 1] vs [competitor 2]

Combined Assessment

Run full 120-item assessment on [domain]: CITE domain audit + CORE-EEAT content audit on [sample pages]

Skill Contract

Gate verdict: TRUSTED (no critical issues, scores above threshold) / CAUTIOUS (issues found but none critical) / UNTRUSTED (a critical trust issue failed — see "Critical Issue to Fix" in the report). Always state the verdict prominently at the top of the report using plain language, not item IDs.

Expected output: a CITE audit report, a citation-trust verdict, and a short handoff summary ready for memory/audits/domain/.

• Reads: the target domain, supporting authority signals, comparison domains, and prior decisions from CLAUDE.md and the shared State Model when available.
• Writes: a user-facing authority report plus a reusable summary that can be stored under memory/audits/domain/.
• Promotes: veto items and domain risks to memory/hot-cache.md (auto-saved). Authority context to memory/audits/domain/. Results feed into entity-optimizer as authority input for brand's canonical profile.
• Next handoff: use the Next Best Skill below once the trust picture is clear.

Data Sources

See CONNECTORS.md for tool category placeholders.

Note: All integrations are optional. This skill works without any API keys — users provide data manually when no tools are connected.

With ~~link database + ~~SEO tool + ~~AI monitor + ~~knowledge graph + ~~brand monitor connected: Automatically pull backlink profiles and link quality metrics from ~~link database, domain authority scores and keyword rankings from ~~SEO tool, AI citation data from ~~AI monitor, entity presence from ~~knowledge graph, and brand mention data from ~~brand monitor.

With manual data only: Ask the user to provide:

1. Domain to evaluate
2. Domain type (if not auto-detectable): Content Publisher, Product & Service, E-commerce, Community & UGC, Tool & Utility, or Authority & Institutional
3. Backlink data: referring domains count, domain authority, top linking domains
4. Traffic estimates (from any SEO tool or SimilarWeb)
5. Competitor domains for comparison (optional)

Proceed with the full 40-item audit using provided data. Note in the output which items could not be fully evaluated due to missing access (e.g., AI citation data, knowledge graph queries, WHOIS history).

Instructions

When a user requests a domain authority audit:

Step 1: Preparation

### Audit Setup

**Domain**: [domain]
**Domain Type**: [auto-detected or user-specified]
**Dimension Weights**: [from domain-type weight table below]

#### Domain-Type Weight Table

> Canonical source: `references/cite-domain-rating.md`. This inline copy is for convenience.

| Dim | Default | Content Publisher | Product & Service | E-commerce | Community & UGC | Tool & Utility | Authority & Institutional |
|-----|:-------:|:-:|:-:|:-:|:-:|:-:|:-:|
| C | 35% | **40%** | 25% | 20% | 35% | 25% | **45%** |
| I | 20% | 15% | **30%** | 20% | 10% | **30%** | 20% |
| T | 25% | 20% | 25% | **35%** | 25% | 25% | 20% |
| E | 20% | 25% | 20% | 25% | **30%** | 20% | 15% |

#### Critical Trust Check (Emergency Brake)

| Check | Status | Action |
|-------|--------|--------|
| Link profile matches real traffic | ✅ Pass / ⚠️ CRITICAL | [If CRITICAL: "Audit backlink profile; disavow toxic links"] |
| Backlink profile is unique to this domain | ✅ Pass / ⚠️ CRITICAL | [If CRITICAL: "Flag as manipulation network; investigate link sources"] |
| No Google penalties or deindexing | ✅ Pass / ⚠️ CRITICAL | [If CRITICAL: "Address penalty first; all other optimization is futile"] |

If any critical trust check triggers, flag it prominently at the top of the report using plain language. CITE Score is capped per Runbook §2.

Step 2: C + I Audit (20 items)

Evaluate each item against the criteria in references/cite-domain-rating.md.

Score each item:

• Pass = 10 points (fully meets criteria)
• Partial = 5 points (partially meets criteria)
• Fail = 0 points (does not meet criteria)

### C — Citation

| ID | Check Item | Score | Notes |
|----|-----------|-------|-------|
| C01 | Referring Domains Volume | Pass/Partial/Fail | [specific observation] |
| C02 | Referring Domains Quality | Pass/Partial/Fail | [specific observation] |
| ... | ... | ... | ... |
| C10 | Link Source Diversity | Pass/Partial/Fail | [specific observation] |

**C Score**: [X]/100

### I — Identity

| ID | Check Item | Score | Notes |
|----|-----------|-------|-------|
| I01 | Knowledge Graph Presence | Pass/Partial/Fail | [specific observation] |
| ... | ... | ... | ... |

**I Score**: [X]/100

Step 3: T + E Audit (20 items)

Same format for Trust and Eminence dimensions.

### T — Trust

| ID | Check Item | Score | Notes |
|----|-----------|-------|-------|
| T01 | Link Profile Naturalness | Pass/Partial/Fail | [specific observation] |
| ... | ... | ... | ... |

**T Score**: [X]/100

### E — Eminence

| ID | Check Item | Score | Notes |
|----|-----------|-------|-------|
| E01 | Organic Search Visibility | Pass/Partial/Fail | [specific observation] |
| ... | ... | ... | ... |

**E Score**: [X]/100

Note: Some items require specialized data (C05-C08 AI citation data, I01 knowledge graph queries, T04-T05 IP/profile analysis). Score what is observable; mark unverifiable items as "N/A — requires [data source]" and exclude from dimension average.

§1 · Handoff Schema (authoritative)

Every auditor-class handoff MUST follow this shape. Emitted audit artifact files (e.g., memory/audits/**/*.md) MUST include class: auditor-output in their YAML frontmatter so the PostToolUse Artifact Gate and Stop-time archiving hooks can detect them by frontmatter class instead of prose pattern-matching. Files lacking this marker are not treated as audit artifacts regardless of body content.

---